import cc.mallet.grmm.inference.Inferencer; //导入依赖的package包/类
private static Inferencer createInferencer (String spec) throws EvalError
{
  String cmd;
  if (spec.indexOf ('(') >= 0) {
    // assume it's Java code, and don't screw with it.
    cmd = spec;
  } else {
    cmd = "new "+spec+"()";
  }

  // Return whatever the Java code says to
  Object inf = interpreter.eval (cmd);

  if (inf instanceof Inferencer)
    return (Inferencer) inf;

  else throw new RuntimeException ("Don't know what to do with inferencer "+inf);
}
 

import cc.mallet.grmm.inference.Inferencer; //导入依赖的package包/类
public void testBadVariable ()
{
  FactorGraph fg = createBoltzmannChain (5);
  Assignment assn = fg.sampleContinuousVars (new Randoms (23423));
  FactorGraph sliced = (FactorGraph) fg.slice (assn);
  Inferencer bp = new TRP ();
  bp.computeMarginals (sliced);

  try {
    bp.lookupMarginal (new Variable (2));
    fail ("Expected exception");
  } catch (IllegalArgumentException e) {
    // expected
    System.out.println ("OK: As expected, got exception "+e);
  }
}
 

import cc.mallet.grmm.inference.Inferencer; //导入依赖的package包/类
public static void reportTrainingLikelihood (ACRF acrf, InstanceList trainingList)
{
  double total = 0;
  Inferencer inf = acrf.getInferencer ();
  for (int i = 0; i < trainingList.size (); i++) {
    Instance inst = trainingList.get (i);
    ACRF.UnrolledGraph unrolled = acrf.unroll (inst);
    inf.computeMarginals (unrolled);
    double lik = inf.lookupLogJoint (unrolled.getAssignment ());
    total += lik;
    logger.info ("...instance "+i+" likelihood = "+lik);
  }
  logger.info ("Unregularized joint likelihood = "+total);
}
 

import cc.mallet.grmm.inference.Inferencer; //导入依赖的package包/类
/**
 * Returns the highest-score Assignment in a model according to a given inferencer.
 * @param mdl Factor graph to use
 * @param inf Inferencer to use.  No need to call <tt>computeMarginals</tt> first.
 * @return An Assignment
 */
public static Assignment bestAssignment (FactorGraph mdl, Inferencer inf)
{
  inf.computeMarginals (mdl);
  int[] outcomes = new int [mdl.numVariables ()];
  for (int i = 0; i < outcomes.length; i++) {
    Variable var = mdl.get (i);
    int best = inf.lookupMarginal (var).argmax ();
    outcomes[i] = best;
  }
  return new Assignment (mdl, outcomes);
}